Recently, with the development and diversification of communication devices, the number of users in a network environment has rapidly increased. In a complex network environment, however, a server may have a functional limitation in processing and responding to messages transmitted from clients. Thus, in order to smoothly perform message interactions with an infinite number of clients, a distributed server system including a plurality of servers has been mostly used.
In order to increase the distribution efficiency of the distributed server system, research has been continuously conducted on a distributed server system as disclosed in Korean Patent Publication No. 10-2000-7009397.
In the conventional network technology, however, when trying to reconnect to a disconnected server, a user must newly go through a user authentication process. Furthermore, even when a user tries to reconnect to the disconnected server using an ID and password which are previously stored therein, an excessive reconnection time may be required by the user authentication process.
Furthermore, when a disconnection occurs, data which are not yet transmitted until a user reconnects to the server after the disconnection may be lost.
Thus, when a disconnection occurs between a server and a client, the conventional network technology not only requires a reconnection process and a re-authentication process, but also requires a separate recovery process for data loss.